Internal combustion engines typically used in motor vehicles generate a substantial amount of noise due to the combustion occurring within the engine. Conventionally, the noise generated is suppressed by a passive muffler system in which the sound waves are broken up by resonance with baffles, passageways and the like or absorbed by fibrous material. However, such techniques of reducing the sound level also obstruct the free flow of exhaust gases through the exhaust conduits and therefore substantially interfere with efficient operation of the vehicle engine by interfering with the release of combustion products and inhibiting the replacement of the combusted gases with fresh fuel in the engine cylinders. Nevertheless, despite the reduction in economy and performance, the need for substantially reduced noise levels requires the use of mufflers on all production motor vehicles.
Although active noise cancellation systems have been employed with large ducts used for heating and ventilation in large buildings, the previously known systems are not well adapted for use in the environment of motor vehicles. For example, U.S. Pat. No. 4,473,906 to Warnaka et al discloses numerous prior art sound attenuation system embodiments. In general, sensed sound pressure produces a signal adapted to drive a loudspeaker for inputting cancellation signals into the duct. The cancellation signal is an acoustic pulse signal 180.degree. out of phase with the signal passing past the speaker through the duct. The prior art embodiments also illustrate improved noise attenuation performance by reducing the effect of the feedback of the cancellation signal which arrives at the sensor. The patent discusses the inclusion of additional transducers and electronic controls to improve the performance of the active acoustic attenuator.
U.S. Pat. No. 4,677,677 to Eriksson further improves attenuation by including an adaptive filter with on-line modeling of the error path and the canceling speaker by using a recurslye algorithm without dedicated off-line pretraining. U.S. Pat. No. 4,677,676 adds a low amplitude, uncorrelated random noise source to a system to improve performance. Likewise, U.S. Pat. Nos. 4,876,722 to Decker et al and 4,783,817 to Hamada et al disclose particular component locations which are performance related and do not adapt active attenuator noise control systems to motor vehicles.
However, none of these improvements render the system applicable to muffle engine noise in the environment of a motor vehicle. For example, such systems often employ extremely large transducers such as 12 or 15 inch loudspeakers of conventional construction. Such components are not well adapted for packaging within the confines of the motor vehicle, and particularly, within the undercarriage of the motor vehicle. Moreover, since the lowest frequency of the signal to be cancelled is on the order of 25 hertz, a large loudspeaker is used to generate sound signals with sufficient amplitude in that range, and such speakers are not practical to mount beneath a motor vehicle. Moreover, although the highest frequencies encountered are easier to dissipate because of their smaller wavelength, the highest frequency to be cancelled is on the order of 250 hertz.
Moreover, many of the above-mentioned systems locate the speakers within the ducts subjected to the sound pressure signal. The loudspeakers conversationally employed in those systems would not fit within conventional exhaust conduits for motor vehicles. Furthermore, the harsh environmental conditions within such a chamber would adversely affect the described known systems and diminish their performance in a motor vehicle.
Although there have been known techniques for increasing the efficiency of audio loudspeakers, those teachings have not been considered readily applicable to active noise attenuating systems. French Patent No. 768,373 to D'Alton, U.S. Pat. No. 4,549,631 to Bose and the Bandpass Loudspeaker Enclosures publication of Geddes and Fawcett presented at the 1986 convention of the Audio Engineering Society acknowledge the phenomena of tuning loudspeaker output by the use of chambers including ports. The recognition of this phenomena has been limited to its effect upon audio reproduction, and particularly dispersion of the audio signal to an open area outside the loudspeaker enclosure. The closed conduit system of motor vehicle exhaust systems, and the harsh conditions associated with such systems, is a substantially different environment.
In addition, my above-identified copending applications discuss improvements and the advantages to be obtained by ported communication of multiple transducer faces with an exhaust conduit. However, the mounting of multiple transducers increases the packaging problems, material costs and assembly complexity of the vehicle. Furthermore, a back to back alignment of transducers may position the magnets so that the magnetic fields can interfere with efficient operation of the transducers.